Level:
Advanced
TCM Section(s):
7.6. Risk Management
3.3. Investment Decision Making
Venue: 2022 AACE International Conference & Expo
Abstract: It is not uncommon for company procedures and project teams in the process industry to directly use the tabulated expected cost accuracy ranges from Recommended Practice (RP) 18R-97 Cost Estimate Classification System – As Applied in Engineering, Procurement, and Construction for the Process Industries in their estimates, with the incorrect presumption that the range is solely determined by the level of scope definition (i.e., the class of the estimate). This article, based on the Rand Corporation (RAND) parametric cost model for risk analysis and contingency determination from RP 43R-08 Cost Estimate Classification System – As Applied in Engineering, Procurement, and Construction for the Process Industries, evaluated model inputs for distinct project scenarios and compared model outcomes with the expected ranges shown in RP 18R-97’s table 1. Built on sample information from the original RAND study (as was done in RP 119R-21, Cost Estimate Accuracy Range and Contingency Determination using Tables Derived from Parametric Risk Models), six scenarios were simulated for scope definition ratings equivalent to FEL 1/Class 5, FEL 2/Class 4 and FEL 3/Class 3, with the manipulation of the model’s other parameters of ‘new technology’, ‘impurity’ and ‘plant complexity’. Results, based solely on these systemic risks, demonstrated frequent RAND model variation from the RP 18R-97 upper ranges when some of the non-definitional risk parameters are reasonably varied. In extreme cases, difficulties were observed in using the Excel NORMINV function to determine the ranges for a parametric model. This paper presents an experimentation of the RAND cost model and demonstrates quantitatively how much model results can diverge from the range-of-ranges tabulated in RP 18R-97.
TCM Section(s):
7.6. Risk Management
3.3. Investment Decision Making
Venue: 2022 AACE International Conference & Expo
Abstract: It is not uncommon for company procedures and project teams in the process industry to directly use the tabulated expected cost accuracy ranges from Recommended Practice (RP) 18R-97 Cost Estimate Classification System – As Applied in Engineering, Procurement, and Construction for the Process Industries in their estimates, with the incorrect presumption that the range is solely determined by the level of scope definition (i.e., the class of the estimate). This article, based on the Rand Corporation (RAND) parametric cost model for risk analysis and contingency determination from RP 43R-08 Cost Estimate Classification System – As Applied in Engineering, Procurement, and Construction for the Process Industries, evaluated model inputs for distinct project scenarios and compared model outcomes with the expected ranges shown in RP 18R-97’s table 1. Built on sample information from the original RAND study (as was done in RP 119R-21, Cost Estimate Accuracy Range and Contingency Determination using Tables Derived from Parametric Risk Models), six scenarios were simulated for scope definition ratings equivalent to FEL 1/Class 5, FEL 2/Class 4 and FEL 3/Class 3, with the manipulation of the model’s other parameters of ‘new technology’, ‘impurity’ and ‘plant complexity’. Results, based solely on these systemic risks, demonstrated frequent RAND model variation from the RP 18R-97 upper ranges when some of the non-definitional risk parameters are reasonably varied. In extreme cases, difficulties were observed in using the Excel NORMINV function to determine the ranges for a parametric model. This paper presents an experimentation of the RAND cost model and demonstrates quantitatively how much model results can diverge from the range-of-ranges tabulated in RP 18R-97.